More than one-third of patients with coronary disease (CAD) have "low" HDL cholesterol (HDLc) levels (less than 35 mg/dl; U.S. 20th percentile) and "normal" LDLc (less than 145; U.S. mean), a group for whom current treatment guidelines are not based on clinical trial data. Low HDLc levels are strong, independent predictors of cardiovascular (CV) disease and CV mortality risk, equally so for both men and women. It is proposed that this high CAD risk is due to an imbalance between delivery of cholesterol into the arterial intima by LDL and its removal by HDL. Also, since HDL serve as antioxidants and cytoprotectants, an important HDL role may be to prevent LDL oxidation and thus limit macrophage- mediated intimal lipid accumulation or to prevent vascular cell toxicity. Recent epidemiologic, experimental, and clinical trial evidence suggests that a 15 mg/dl rise in HDL cholesterol would reduce CAD incidence and mortality by 30-70% and that antioxidant vitamins E, C, and Beta-carotene might reduce CAD events and atherogenesis. The potential absolute benefit is much greater in those with existing CAD. It has also been shown that HDLc rises in response to exercise, smoking cessation, weight reduction and monounsaturated fats. Our goal is to test, in patients with low HDLc normal LDLc, and CAD, two hypotheses that spring from these observations: 1) that reducing the LDLc-HDLc imbalance (ratio) by pharmacological and hygienic means will halt or reverse the progression of atherosclerosis, and 2) that vitamins E, C and Beta-carotene will independently reduce atherogenesis and thus enhance the LDL-HDL treatment effect. Therefore, we propose to randomize 160 such men (less than 62 y.o.) and women (less than 67 y.o.) to double-blinded therapy, assigned in a factorial design: 1) mid-dose niacin (500 mg qid), plus simvastatin (20 mg qd); 2) an antioxidant cocktail (vitamin C (500 mg bid), E (400 IU bid), and Beta-carotene (12.5 mg qd)); 3) a combination of the above two; or 4) double placebo. All these patients will adopt hygienic means for raising HDLc. This will be accomplished by professional counseling in diet, smoking cessation, and a structured exercise program. The primary endpoint in these studies will be CAD progression as assessed in a fully blinded analysis by serial quantitative arteriography. Secondary endpoints include the frequency of cardiovascular events (death, infarction, and revascularization) as well as intimal disease change assessed in a study subset by intravascular ultrasound. Mechanisms of benefit will be assessed in terms of the correlation of disease change or clinical event rate with the therapy-induced changes in lipoproteins (LDL, HDL, HDL2, IDL) or in apoproteins {B, AI, AII, E and (a)} or in dietary variables, or in the plasma levels of antioxidant vitamins, or in the resistance of LDL to oxidation. This 5-year placebo-controlled study employs state-of-the-art techniques to assess both the magnitude and mechanisms of benefit. It addresses the question of effective therapy for the roughly one-third of patients for whom CAD appears associated principally with low HDLc, among whom there has not been a directed clinical or angiographic trial, and among whom appropriate therapy is currently controversial. As such, it has great potential public health impact.